Genetic polymorphism of CYP2C19 plays an important role in drug response. About 10% of the current commonly used drugs are metabolized by the CYP2C19 enzymes (Zhou et al. 2009). To date, more than 39 (CYP2C19*1-*39) allelic variants and subvariants of CYP2C19 have been reported (https://www.pharmvar.org/gene/CYP2C19). CYP2C19*2 (rs4244285, c.681G > A; splicing defect) and CYP2C19*3 (rs4986893, c.636G > A; premature stop codon), which encode deficient drug-metabolizing enzymes, are the two major mutant alleles of CYP2C19 in the Asian population (Zhou et al. 2009; Lee et al. 2018). In turn, the CYP2C19*10 allele (rs6413438, c.680C > T; p.Pro227Leu) is located 1 bp upstream of the null CYP2C19*2 allele (Blaisdell et al. 2002) and causes decreased enzyme activity in vitro (Blaisdell et al. 2002; Wang et al. 2011a, 2011b; Langaee et al. 2014; Takahashi et al. 2015). Moreover, the CYP2C19*17 allele (rs12248560, c.−806C > T) is a mutation present in the 5′-flanking region that increases gene transcription and CYP2C19 activity in vivo (Sim et al. 2006). Pharmacokinetic changes caused by genetic polymorphism of the gene encoding the CYP2C19 enzyme have been reported for various drugs (Scott et al. 2012; Choi et al. 2014; Lee et al. 2018; Cho et al. 2023).

Two cases of miscalling of the CYP2C19*10 allele as the CYP2C19*2 allele have been reported (Rasmussen and Werge 2008; Langaee et al. 2014). One case was analyzed using the polymerase chain reaction-restriction fragment length polymorphism (PCR-RFLP) assay (Rasmussen and Werge 2008), and the other case was analyzed using the TaqMan assay (Langaee et al. 2014). The PCR-RFLP method using SmaI has been widely employed for the analysis of the CYP2C19*2 allele (De Morais et al. 1994a; de Morais et al. 1994b; Goldstein and Blaisdell 1996; Daly et al. 2006). In the CYP2C19*2 variant, the restriction enzyme recognition site of SmaI is abolished. SmaI recognizes the sequence from c.678C to c.683G of CYP2C19*1. Therefore, the occurrence of mutation in the region of the recognition sequence does not allow their distinction from each other. CYP2C19*10 (c.680C > T) is a mutation that differs by 1 bp from CYP2C19*2 (c.681G > A) and cannot be distinguished by the conventional PCR-RFLP method using SmaI. Thus, the frequency of CYP2C19*10 may appear high in an ethnic group with a high frequency of CYP2C19*2. We classified CYP2C19*2 and *10 using the pyrosequencing method and performed a study to accurately confirm their frequency.

Subjects

A total of 432 healthy Korean volunteers were recruited into the study and genotyped for the CYP2C19 polymorphisms. Written informed consent was obtained from each subject before participation in the study. All procedures were approved by the Institutional Review Board of Keimyung University (Daegu, Republic of Korea, approval number: 40525-201509-BR-70-02) and were performed in accordance with the recommendations of the Declaration of Helsinki on biomedical research involving human subjects.

DNA extraction and PCR

A 3-mL blood sample was obtained from each subject, and deoxyribonucleic acid (DNA) was isolated using an extraction kit (Wizard Genomic DNA Purification Kit; Promega, Madison, WI, USA). The genotyping of CYP2C19*2 (c.681G > A), *3 (c.636G > A), *10 (c.680C > T), and *17 (c.−806C > T) was carried out using pyrosequencing. PCR was performed in a final volume of 20 μL consisting of genomic DNA (200 ng), 20 pM each primer, and PCR mixture (250 μM dNTPs, 10 mM Tris-HCl buffer (pH 9.0), 30 mM KCl, 1.5 mM MgCl₂, and 1U Taq DNA polymerase) (AccuPower PCR premix, Bioneer, Korea). Amplification was performed using a T100 Thermal cycler (Bio-Rad, Hercules CA, USA).

Pyrosequencing assay

The primers used for PCR and pyrosequencing were designed using the PyroMark Assay Design software 2.0 (Qiagen, Hilden, Germany). The primers used for PCR for CYP2C19 pyrosequencing are shown in Table 1. The PCR thermal cycling conditions consisted of denaturation at 94℃ for 5 min; followed by 35 cycles of denaturation at 94℃ for 30 s, annealing at 60℃ (for CYP2C19*2, *10, and *17) or 58℃ (for CYP2C19*3) for 30 s, and extension at 72℃ for 30 s; and a final extension at 72℃ for 5 min. The target-gene fragment was amplified using PCR (T100 Thermal cycler, Bio-Rad), and the PCR products were sequenced by pyrosequencing. All pyrosequencing reactions were performed on a PyroMark Q96 ID instrument using PyroMark Gold Q96 reagents, according to the manufacturer’s protocols (Qiagen, Hilden, Germany).

Table 1 . Oligonucleotide primers used for the PCR and pyrosequencing assays.

Allele	PCR conditions					Pyrosequencing conditions
	PCR primer	Annealing temperature (℃)	Number of cycles	PCR product size (bp)		Sequencing primer	Sequence to analyze
CYP2C19*2 and *10	Forward primer: B5’-CCAGAGCTTGGCATATTGTATCTA-3’ Reverse primer: 5’-CGCAAGCAGTCACATAACTAAGC-3’	60	35	289		5’-AAGTAATTTGTTATGGGTTC-3	C[C/T][G/A]GGAAATAATCAATG
CYP2C19*3	Forward primer: 5’-ATTGAATGAAAACATCAGGATTGTA-3’ Reverse primer: B5’-TTCCCAGAAAAAAAGACTGTAAGTG-3’	58	35	99		5’-TTGTAAGCACCCCCT-3’	G[G/A]ATCCAGGTAA
CYP2C19*17	Forward primer: 5’-GTGATGGAGAAGGGAGAACTCTTA-3’ Reverse primer: B5’-CATCGTGGCGCATTATCTCTTA-3’	60	35	254		5’-TTGTGTCTTCTGTTCTCAA-3’	AG[C/T]ATCTCTG

B, biotinylated on 5’-end of primer..

Briefly, 20 μL of amplified DNA were mixed with 1.5 μL of Streptavidin Sepharose High-Performance beads (Cytiva, Emeryville, CA, USA) and 58.5 μL of binding buffer, followed by shaking at 1,400 rpm for 10 min. Subsequently, the immobilized biotinylated PCR product/streptavidin high-performance bead complex was captured using a Vacuum Prep Workstation (Qiagen). Single-strand purification was achieved by washing the Vacuum Prep Workstation sequentially with 70% ethanol, denaturation solution, and wash buffer, respectively, twice (for 30 s each time). In this step, the unbiotinylated strand was dissociated and discarded. The beads that were bound to single biotinylated strands were released to a PSQ 96 Plate Low (Qiagen), to which 40 μL annealing buffer and 1.2 μL complementary sequencing primer had been previously added. The PSQ 96 Plate Low was incubated at 85℃ for 2 min, followed by slow cooling to room temperature. The processed mixture was loaded onto the PyroMark Q96ID system (Qiagen) equipped with the PyroMark Q96 software (Qiagen), for pyrosequencing.

Statistical analysis

Data were compiled according to the genotype and allele frequencies. The frequencies of each allele were reported with 95% confidence intervals. Hardy–Weinberg equilibrium was evaluated by comparing the genotype frequencies with the expected values using a contingency table χ² test. Statistical significance was determined using the χ² test and set at p < 0.05.

A representative chromatogram of direct DNA sequencing and the pyrosequencing program used for CYP2C19 genotyping are shown in Figs 1-3. The pyrosequencing-based genotyping results corresponded well to the results of direct DNA sequencing in randomly selected samples using the same genomic DNA.

Figure 1. Direct DNA sequencing chromatograms (A) and pyrosequencing pyrograms (B) for the CYP2C19*2 and *10 alleles. Chromatograms are shown for each genotype sequence in the reverse direction. The order of nucleotide dispensation used for pyrosequencing is shown under the pyrograms. The chromatograms and pyrograms presented here are from the same three patients. The shaded area on the left indicates the c.681G > A locus, and the shaded area on the right indicates the c.680C > T locus.

Figure 2. Direct DNA sequencing chromatograms (A) and pyrosequencing pyrograms (B) for the CYP2C19*3 allele. The order of nucleotide dispensation used for pyrosequencing is shown under the pyrograms. The chromatograms and pyrograms presented here are from the same three patients. The shaded area indicates the c.636G > A locus.

Figure 3. Direct DNA sequencing chromatograms (A) and pyrosequencing pyrograms (B) for the CYP2C19*17 allele. The order of nucleotide dispensation used for pyrosequencing is shown under the pyrograms. The chromatograms and pyrograms presented here are from the same three patients. The shaded area indicates the c.−806C > T locus

In this study, the frequency of the CYP2C19 alleles was assessed in 432 Korean subjects who were not related to each other. The estimated frequencies of the CYP2C19 alleles and genotypes in the Korean population are summarized in Table 2. The genotype frequency distribution did not deviate significantly from the Hardy–Weinberg equilibrium.

Table 2 . CYP2C19 allele (A) and genotype (B) frequencies in a Korean sample. The expected genotype frequencies were calculated from the allele frequencies using the Hardy–Weinberg equation.

(A)
Allele	n (864)	Frequency (%)	95% CI
CYP2C19*1	544	63.0	59.7, 66.3
CYP2C19*2	228	26.4	23.4, 29.4
CYP2C19*3	81	9.4	7.4, 11.4
CYP2C19*10	0	0.0	0.0, 0.0
CYP2C19*17	11	1.3	0.5, 2.1
(B)
Genotype	Number of subjects (432)	Observed frequency (%)	95% CI	Expected frequency (%)
CYP2C19*1/*1	173	40.0	35.3, 44.8	39.6
CYP2C19*1/*17	7	1.6	0.4, 2.8	1.6
CYP2C19*17/*2	3	0.7	−0.1, 1.5	0.7
CYP2C19*17/*3	1	0.2	−0.2, 0.7	0.2
CYP2C19*1/*2	143	33.1	28.6, 37.6	33.2
CYP2C19*1/*3	48	11.1	8.1, 14.1	11.8
CYP2C19*2/*2	30	6.9	4.5, 9.4	7.0
CYP2C19*2/*3	22	5.1	3.0, 7.2	4.9
CYP2C19*3/*3	5	1.2	0.1, 2.2	0.9

In the 432 Koreans studied here, the functional CYP2C19*1 allele was present at a frequency of 63.0% (95% CI: 59.7, 66.3). In cases of two non-functional alleles, CYP2C19*2 was the most common variant allele at 26.4% (95% CI: 23.4, 29.4), whereas the CYP2C19*3 allele was present in 9.4% (95% CI: 7.4, 11.4) of the subjects. The CYP2C19*17 allele, increased function allele, was present at a frequency of 1.3% (95% CI: 0.5, 2.1).

Among the cohort of 432 Koreans, the CYP2C19*1/*1 genotype, which yields normal enzyme activity and a normal metabolizer (NM) phenotype, was present in 40.0% of the individuals (95% CI: 35.3, 44.8). The number of subjects with the CYP2C19*2/*2, *2/*3, and *3/*3 genotypes, who were classified as poor metabolizers (PMs), was 30, 22, and 5, respectively; therefore, the frequency of CYP2C19 PMs in the Korean population was about 13.2%. The CYP2C19*1/*2, *1/*3, *17/*2, and *17/*3 genotypes, which yield decreased enzyme activity and an intermediate metabolizer (IM) phenotype, were present at frequencies of 33.2% (95% CI: 28.6, 37.6), 11.8% (95% CI: 8.1, 14.1), 0.7% (95% CI: −0.1, 1.5), and 0.2% (95% CI: −0.2, 0.7), respectively. Therefore, the frequency of CYP2C19 IMs in the Korean population was about 45.1%. The CYP2C19*1/*17 genotype, which is classified as a rapid metabolizer (RM), was present in 1.6% (95% CI: 0.4, 2.8) of the individuals in our cohort. Finally, none of the subjects carried the CYP2C19*10 allele or were homozygous for the CYP2C19*17 allele.

Although reports of the frequency of the CYP2C19*10 allele are scarce, it exhibited a low frequency of less than 1% (Table 3) (Nakamoto et al. 2007; Rasmussen and Werge 2008; Langaee et al. 2014; Khalil et al. 2016; Kim et al. 2021; Goljan et al. 2022).

Table 3 . Comparison of the reported CYP2C19*10 allele frequencies among various ethnic groups.

Populations	Number of subjects	CYP2C19*10 allele frequency (%)	Reference
Korean	432	0.00	Present study
Korean	1,012	0.00	Kim et al. 2021
Saudi	11,889	0.00	Goljan et al. 2022
Egyptian	190	0.00	Khalil et al. 2016
European Caucasian	414	0.12	Rasmussen and Werge 2008
European Caucasian	80	0.00	Nakamoto et al. 2007
African-American	181	0.80	Langaee et al. 2014
Hispanics	202	0.25	Langaee et al. 2014

The CYP2C19*10 allele involves a c.680C > T transition in exon 5 of the CYP2C19 gene, leading to a p.Pro227Leu substitution in the encoded protein (Blaisdell et al. 2002). Proline 227 is located within the F-G loop, thus being part of the substrate access channel in CYP2C19, which plays an important role in the substrate specificity and activity of the enzyme (Tsao et al. 2001; Wada et al. 2008). Although different from the results of in vitro studies of CYP2C19 substrates, the CYP2C19*10 variant appeared to have lower activity than did the CYP2C19*1 allele. The intrinsic clearance values for CYP2C19*10, as assessed based on the levels of the CYP2C19 protein, were decreased to 3%-43% of the value obtained for CYP2C19*1 (Blaisdell et al. 2002; Wang et al. 2011a, 2011b; Langaee et al. 2014; Takahashi et al. 2015). Moreover, the CYP2C19*10 allelic variant showed substrate specificity. The impact of CYP2C19*10 on the metabolism of various substrates could not be precisely defined; therefore, further studies using substrates of interest are needed.

The CYP2C19*10 allele was not detected in the Korean population studied here (Table 3). Other populations also showed a low frequency of this allele (< 1%) (Table 3). To date, no data have been reported regarding the in vivo activity of CYP2C19*10 in humans. Because the frequency of the CYP2C19*10 allele is very low in general, the accurate evaluation of its clinical function seems to be difficult.

The presence of the CYP2C19*10 allele in close proximity to the CYP2C19*2 allele interferes with the CYP2C19*2 genotyping assay. Two cases of misclassification of the CYP2C19*10 allele as the CYP2C19*2 allele have been reported (Rasmussen and Werge 2008; Langaee et al. 2014). In one case, CYP2C19*1/*10 was misclassified as CYP2C19*1/*2 by PCR-RFLP using the SmaI assay (Rasmussen and Werge 2008). In the other case, CYP2C19*10/*2 was miscalled as CYP2C19*2/*2 based on the results of a TaqMan genotyping assay (Langaee et al. 2014). Therefore, methods such as pyrosequencing, which can confirm the DNA sequence directly, will be useful for genotyping the CYP2C19*10 allele.

In the present study, we observed to following allele frequencies: 63.0% for CYP2C19*1, 26.4% for CYP2C19*2, 9.4% for CYP2C19*3, and 1.3% for CYP2C19*17. The frequency of the CYP2C19*17 allele was similar to that reported by other Korean (0.3%-1.5%) or Asian (0.5%-3%) studies, and was lower than that reported for Caucasians (15.4%-33.8%) or Africans (10%-18%) (Sim et al. 2006; Kim et al. 2010; Li-Wan-Po et al. 2010; Ramsjö et al. 2010; Scott et al. 2011; Kim et al. 2021). Based on the ability to metabolize CYP2C19 substrates, individuals are categorized into the five phenotype categories recommended by the Clinical Pharmacogenetics Implementation Consortium (CPIC), as follows: PMs, IMs, NMs; formerly known as extensive metabolizers (EM), RMs, and ultrarapid metabolizers (Caudle et al. 2017; Lee et al. 2022). The frequencies of the CYP2C19 NM, RM, IM, and PM phenotypes among the 432 Koreans were 40.0%, 1.6%, 45.1%, and 13.2%, respectively. Moreover, a higher incidence of CYP2C19 PMs was found among Koreans (13%-16%) compared with Caucasians and Africans (1%-7%) (De Morais et al. 1994a; de Morais et al. 1994b; Roh et al. 1996; Zhou et al. 2009; Kim et al. 2010; Ramsjö et al. 2010).

In summary, the frequencies of the CYP2C19*2, *3, and *17 alleles in the Korean population were estimated to be 26.4%, 9.4%, and 1.3%, respectively. The frequencies of the CYP2C19*2 and *3 alleles were relatively high, whereas the frequency of the CYP2C19*17 allele was low. Finally, none of the subjects carried the CYP2C19*10 allele. The frequency of the CYP2C19*10 allele seems to be extremely low in the Korean population.

This research was supported by the Basic Science Research Program through the National Research Foundation of Korea (NRF) funded by the Ministry of Science, ICT, and Future Planning (NRF-2017R1D1A3B04034852).

The authors declare that they have no conflict of interest.